English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The localization of objects is a crucial task in various applications such asrobotics, virtual and augmented reality, and the transportation of goods inwarehouses. Recent advances in deep learning have enabled the localizationusing monocular visual cameras. While structure from motion (SfM) predicts theabsolute pose from a point cloud, absolute pose regression (APR) methods learna semantic understanding of the environment through neural networks. However,both fields face challenges caused by the environment such as motion blur,lighting changes, repetitive patterns, and feature-less structures. This studyaims to address these challenges by incorporating additional information andregularizing the absolute pose using relative pose regression (RPR) methods.RPR methods suffer under different challenges, i.e., motion blur. The opticalflow between consecutive images is computed using the Lucas-Kanade algorithm,and the relative pose is predicted using an auxiliary small recurrentconvolutional network. The fusion of absolute and relative poses is a complextask due to the mismatch between the global and local coordinate systems.State-of-the-art methods fusing absolute and relative poses use pose graphoptimization (PGO) to regularize the absolute pose predictions using relativeposes. In this work, we propose recurrent fusion networks to optimally alignabsolute and relative pose predictions to improve the absolute pose prediction.We evaluate eight different recurrent units and construct a simulationenvironment to pre-train the APR and RPR networks for better generalizedtraining. Additionally, we record a large database of different scenarios in achallenging large-scale indoor environment that mimics a warehouse withtransportation robots. We conduct hyperparameter searches and experiments toshow the effectiveness of our recurrent fusion method compared to PGO.

Fusing Structure from Motion and Simulation-Augmented Pose Regression from Optical Flow for Challenging Indoor Environments